Lipid Biomarkers, Carbon Isotopes, and Phylogenetic Characterization of Bacteria in California and Nevada Hot Springs

Abstract

Microbial mats were collected from hot springs in California (Eagleville) and Nevada (Paradise Valley and Crescent Valley) to determine bacterial community structure and pathways of carbon cycling in different geothermal environments of the western United States. Phospholipid fatty acids (PLFA) at Eagleville contained even-numbered fatty acids, with 16:0 being the most abundant (48.8%), followed by 18:1ω 9c (17.2%), 16:1ω 7c/t (6.3%), and 18:0 (6.2%), which are consistent with lipid profiles of cyanobacteria or other phototrophic bacteria. The PLFA profiles at Paradise Valley and Crescent Valley were dominated by similar even-numbered fatty acids; however, branched fatty acids such as iso- and anteiso- 15:0 and 17:0 were also abundant (up to 7.1% compared to 2.0% at Eagleville), suggesting greater relative abundance of heterotrophic bacteria in these springs. Analysis of neutral lipids was only performed on Eagleville and Paradise Valley springs, which revealed abundant bacterial hopanoids including the 2–methylbacteriohopane-32,33,34,35-tetrol (2-methylBHT) that is specific to cyanobacteria; however, the diversity of hopanoid compounds was significantly lower at Eagleville than at Paradise Valley. The carbon-isotope composition of individual PLFA averaged −30.7 ± 1.3‰ (n = 7) at Eagleville, −28.0 ± 1.8‰ (n = 3) at Crescent Valley, and −29.7 ± 3.1‰ (n = 12) at Paradise Valley. Carbon isotope fractionation between PLFA and CO ₂ was only available for Eagleville (−11.7‰) and Paradise Valley (−21.7‰), which indicated the predominance of the Calvin cycle for CO ₂ fixation in these hot springs. Bacterial 16S rRNA genes were extracted from environmental samples at Eagleville and Paradise Valley but not Crescent Valley. Clone libraries indicated the predominance of cyanobacteria (50–75%) at these locations, which is consistent with the lipid profiles. Phylogenetic tree of the 16S rRNA genes indicated that most of the cyanobacterial sequences are unknown and may be specific to the Nevada and California hot springs. Phototrophic green non-sulfur bacteria were also present at Eagleville (13%) and Paradise Valley (7%). The remaining sequences were related to α-, β -, and γ -Proteobacteria, Acidobacteria, Deinococcus/Thermus, Bacteroidetes, and Spirochaetes. However, not all of these sequences were present at each of the springs. Results of this study demonstrate the consistency among lipid profiles (phenotypes), carbon isotopes (biogeochemistry), and 16S rRNA genes (genotypes) of the bacterial community in these hot springs, which cumulatively suggest the importance of cyanobacteria in primary production of biomass under the environmental conditions examined.

Keywords:

• Bacteria
• lipid biomarkers
• hopanoids
• carbon isotopes
• 16S rRNA genes
• cyanobacteria
• California and Nevada hot springs

Acknowledgments

Current address for Yi-Liang Li is Center for Biomarker Analysis, The University of Tennessee, Knoxville, TN 37932, USA.

J. Hutchings at the Eureka County Natural Resources of Nevada helped with sampling hot springs in Paradise Valley and Crescent Valley. The owners of the Surprise Valley showed great hospitality and supported our sampling efforts. J. Cantu and A. Peacock helped with lipid extraction. Deno Karapatakis helped with the art work of Figure 1. This research was supported by the National Science Foundation grant MCB-0348180 (CLZ, CSR, JW, GM), and the U.S. Department of Energy under Award Number DE-FC09-07SR22506 to the University of Georgia Research Foundation (CLZ, CSR, GM). HT and RAG thank the UK Natural Environmental Research Council (NERC) for funding (RAG) and The Science Research Infrastructure Fund (SRIF) from HEFCE for funding the purchase of the ThermoFinnigan LCQ ion trap mass spectrometer. Mr Bernie Bowler and Mr Paul Donohoe (Newcastle University) are thanked for assistance with the mass spectrometers.

Notes

* Calculated from carbon isotope of dissolved inorganic carbon according to Zhang et al. (2004) using the equations of δ¹³C_CO2 = ϵ_{(CO2 - HCO3)} × (1000 + δ¹³C_HCO3)/1000 + δ¹³C_HCO3 and ϵ_{(CO2 - HCO3)} = 24.12 − 9866/T (Mook et al. 1974), where ϵ is the fractionation between dissolved CO₂ and HCO₃ ⁻ and T is absolute temperature of spring water. δ¹³C_HCO3≈ δ¹³C_DIC.

^‡Total organic carbon.

^†Total bacteria (B_m) was calculated using the formula: B_m = (PLFA/(5 × 10⁴))/10^{− 12}, where B_m is total bacterial biomass in cells/g sediment and 1 cell = 5.0 × 10⁴ nmole PLFA = 10^{− 12} g (Chapelle 2001).

* Peak numbers 1–14 are from GC-MS chromatographs in Figure 3; peak numbers 15–19 are from LC-MS chromatographs in Figure 4.

^‡Total hopanoids is sum of total tetrafunctionalized components including BHT, plus total penta-and hexafunctionalized (including all C-2 methylated) components as measured using periodic acid method (see text).

^†Detected in a sample collected in 2005.